France on the Edge of a Climate Precipice: Droughts, Heatwaves, and the Collapse Thresholds of Ecosystems

In June 2026, France is living through what Météo-France calls a “historic heatwave”: the most intense ever recorded at the national level to date, surpassing the records of 2003, with a day–night average of 30 °C on 24 June and 72 départements simultaneously on red alert. This heatwave is part of a climate trajectory clearly mapped out by scientific scenarios: metropolitan France is undergoing structural warming of about +2 °C by 2030 (relative to the pre-industrial era), +2.7 °C by 2050, and up to +4 °C by 2100 according to the TRACC reference trajectory adopted by the French government in 2023. This report explores the climatic, hydrological, and biological conditions that lead to the tipping points of French vegetation, forests, and agricultural systems, and attempts to answer the essential question: how far, and for how long, can we go before collapse?1234

1. The immediate situation (June 2026)

The record heatwave of June 2026

France experienced two major heatwaves even before the official start of summer 2026. The first, in late May (21–30 May), was unprecedented for its earliness and its intensity: temperatures never before reached at that point in the year, more than half the country recording monthly heat records across a panel of some 600 stations although no absolute record fell, and heat persisting over the whole territory for eight consecutive days. It also brought the first heatwave alert ever triggered in May since the vigilance system was created in 2004. Spring 2026 (March–May) showed a national anomaly of +1.7 °C relative to the 1991–2020 normal, making it the hottest spring since 1930, with a rainfall deficit of −23%.56

The second wave, which began around 13 June and was still ongoing at the end of June, surpassed all records. On 24 June 2026, the national average temperature (day and night combined) reached 30 °C, an unprecedented level in French meteorological records, and a record 58 départements were on red alert, rising to 72 the following day. Exceptional maxima were recorded, among them 43.8 °C at Pulluau (Vendée). According to Météo-France, this episode is “at least equivalent to that of 2003,” which had caused 15,000 deaths in France.374

State of soils and groundwater

What the media underplay is the state of soils and underground reserves. According to the BRGM (Bureau de recherches géologiques et minières), as of 1 June 2026, 77% of aquifers showed declining levels, the result of the persistent rainfall deficit since mid-May. By mid-June, the situation had worsened further: 86% of levels were declining, a direct consequence of the lack of effective rainfall. The downward trend was expected to continue, particularly for the most reactive aquifers of the Limousin, the Armorican Massif, and the north-eastern Jurassic arc of the Paris Basin.89

2. Climate trajectories, 2030–2100

Temperature and precipitation projections

France is warming faster than the global average. Based on the TRACC trajectory (the Reference Warming Trajectory for Adaptation to Climate Change), the median projections for metropolitan France are:1

These figures are national averages. Regional gradients are significant: the Mediterranean seaboard, the Aquitaine basin, and the Massif Central warm faster and dry out more than Brittany or the Hauts-de-France. Concretely, by 2050 Rennes will have the current climate of Bordeaux and Marseille the summers of Barcelona, and by the end of the century the Paris region could have the current climate of Montpellier.102

The acceleration of extremes

The most worrying element is not the rise in averages, but the explosion in the frequency of extremes. A 2026 INRAE study in Agricultural and Forest Meteorology quantifies this for French wheat: under a high-emissions scenario, extreme events that historically occurred about once a decade become common by 2100, and often occur together. Under a low-emissions scenario aligned with the Paris Agreement, the same study finds climate risks close to current conditions.11

• The co-occurrence of heatwave + drought could become 3 times more frequent in northern France and 6 times more frequent in the south by 2100.
• Mild winters combined with very wet springs (favourable to diseases and pests) could increase 12-fold in northern France.
• New, unprecedented climate risks could emerge for wheat, such as early heat stress and overly warm nights during grain maturation.

In terms of heatwaves, without strong action France could by 2050 experience heat events 80% more frequent and lasting roughly fifteen times as long as they do today.12

Soil drought: the time bomb

The most critical parameter for vegetation is soil moisture. French projections show that by 2050, France could experience up to 120 days of soil drought per year along the Mediterranean coast. Even more telling, Météo-France projections published by the SDES find that average soil moisture at the end of the century could correspond to today’s extreme dry conditions. In other words, what is today a rare event would become the norm. The dry period would lengthen by 2 to 4 months, at the expense of the wet period.131

3. Vegetation collapse thresholds: what the science says

The concept of a tipping threshold

Ecological science distinguishes two phases in vegetation’s response to drought. In Phase A (stable resistance), the coincidence between drought and vegetation decline stays low and changes little: vegetation withstands slight to mild soil water deficits, drawing on deeper soil water that shallower roots cannot reach. In Phase B (rapid response), once the drought crosses a critical threshold (the TSMsurf, or threshold of soil moisture for vegetation response), the coincidence between water deficit and vegetation decline rises abruptly. This is the inflection point at which resistance gives way.14

This pattern is not marginal: 82 to 86% of the world’s vegetated areas show this kind of non-linear response. Forests, whose woody structure and deep roots let them cope better, have slightly lower thresholds (more resistant) than grasses, in tropical, temperate and boreal regions alike, but they are not immune.14

Drought indices as warning indicators

The SPEI (Standardized Precipitation-Evapotranspiration Index) is among the most widely used measures of drought severity. Studies of the 2018 droughts in the Netherlands and Belgium show that the early-warning threshold for most vegetation types corresponds to an SPEI value of about −1; beyond it, mosaic grasslands and tree or shrub croplands such as orchards are the most vulnerable, and broadleaf forests react faster than other forest types while mixed forests suffer less.15

Studies of flash droughts (rapid droughts), which have become more frequent over 74% of global land since 1951, indicate that vegetation loss there is 1.5 times higher than in conventional droughts, with an NDVI decline of 9.0% versus 5.3%. Recovery to pre-drought vegetation levels takes about 31 pentads on average, roughly five months, against 27.8 after a conventional drought.1617

Duration and intensity: critical temporal thresholds

Research on mixed conifer forests in California (2012–2015) showed that four years of precipitation 45% below the historical normal led to a roughly 75% reduction in gross primary productivity, signalling a potential tipping point. Adding only 157 mm of water in a particularly dry year was enough to steer the ecosystem toward recovery instead of collapse. The response is non-linear and highly sensitive to boundary conditions.18

For European cereal crops, flash droughts, meaning soil-moisture deficits that build over a few weeks to a couple of months, are most frequent during the planting and pollination window and leave most crops water-stressed with reduced yields. Temperatures above 30 °C for three consecutive days or more during wheat flowering can cause pollen sterility of 10 to 30%. For Western Mediterranean vegetation, a simultaneous drought and heatwave multiplies median burned area by 2.1, fire radiative power by 2.9, and fire duration by 1.4. The biomass loss is driven by the drought and not by the combination: about −10% leaf area index on average, reaching −23% in the worst-hit areas, against about −3% for a heatwave alone.192021

4. The forest collapse already under way

Mortality already doubled, or more

French forests are not anticipating a collapse: they are already living through an acute crisis, documented with figures by the IGN. The 2025 National Forest Inventory is alarming: tree mortality has increased by 125% in ten years (2015–2023 vs 2005–2013); 16.7 million m³ of dead trees per year on average, versus 7.4 million over 2005–2013; 8% of the 2.3 billion trees inventoried show signs of deterioration; the volume of standing dead wood has doubled in ten years, reaching 159 million m³; and tree growth has slowed by 4%.222324

Most worrying: in some north-eastern regions (Vosges, Jura, Ardennes), the flux balance is now negative: the volume of dead or harvested wood exceeds natural growth, and the stock of living wood is declining. The most affected species are common spruce (decimated by bark beetles), ash (chalara dieback), chestnut, and increasingly pedunculate oak, of which 10% now show signs of deterioration.2322

Beech, an emblematic species on the brink

Fagus sylvatica, the common beech, is perhaps the most telling indicator of the tipping point under way. Following the exceptional 2018 drought, a study in Plant Biology monitored 963 beeches across three regions of northern Switzerland from 2018 to 2021. Trees that suffered premature leaf senescence in 2018 experienced cumulative mortality of 7.2% by 2021, versus 1.3% for those with normal senescence; their mean crown dieback peaked at 29.2% in 2020, against 8.1% for the trees with normal senescence; and the first signs of recovery appeared only in 2021, three years after the event.25

Similar results have been documented in north-eastern France (SILVA, INRAE): the 2018–2020 episode induced an unprecedented decline of beech in Central Europe, and “beech trees may have reached a tipping point where many individuals are no longer able to survive.” The sequence of 2003 and 2018–2019 in under 20 years shows the destabilising effect of repeated episodes: the second drought strikes trees still weakened by the first.2627

The paradox of the vanishing carbon sink

French forests sequestered on average 63 MtCO₂ per year over 2005–2013. Over 2015–2023, this fell to 39 MtCO₂ per year, a drop of 38%. The IGN still counts the forest as a major climate asset nationally, but in some massifs, notably the North-East, mortality and sanitary harvests already exceed biological growth and the stock is being drawn down. A stock being drawn down is a source. This positive feedback loop (less sink → more CO₂ → more warming → less sink) is one of the most concrete examples of a systemic tipping point.22

5. French agriculture facing collapse

Recent precedents as warnings

The 2003 heatwave remains the historical benchmark for what a severe heatwave means for French agriculture: grain maize −30% versus 2002; fruit −25%; wheat (less affected because it had nearly reached maturity) −21%; forage −30% on the national average; uninsured farm-sector losses of around 4 billion euros for France, the largest national share of the 13 billion estimated across the European Union. Those impacts resulted from a heatwave of a few weeks. The 2026 situation (an early May heatwave plus a second June wave on already dried-out soils) is considered by Météo-France at least equivalent in severity to 2003.283

Medium-term projections: the convergence of risks

A 2025 foresight study by Bureau T, the strategy unit of the SCET group (a Caisse des Dépôts subsidiary), found that 16 of the 24 crops analysed are strongly vulnerable to climate change; the most exposed are arboriculture, maize, and market gardening; and 42% of France’s utilised agricultural area, across 54 départements, is threatened by 2050, in what the study calls a “conjunction of crises”.29

The European Commission (JRC PESETA IV) projects that under moderate-to-high warming without irrigation, a near-collapse of European maize production is projected around 2050, with yield decreases above 23% in all EU countries and above 80% in Member States such as Portugal, Bulgaria, Greece, and Spain. The report places France in its Northern Europe group, where irrigated maize declines by 1 to 14% (against 4 to 22% in Southern Europe), and where it projects wheat gains of 5 to 16%; the wheat losses of up to 49% belong to Southern Europe, not to France. The Haut Conseil pour le Climat is blunt about the direction of travel: climate change is already affecting agricultural production “with major crop losses and reduced feed resources for livestock”, and at around +2 °C by 2050, without further adaptation, it expects additional losses for summer crops such as maize, with fodder production falling in lowland regions and the Southern Alps even as it rises in the coldest areas.3031

The water constraint: the ultimate limiting factor

Irrigation is presented as an adaptation solution, but it runs into an insurmountable contradiction: where drought intensifies, water resources decline. By 2050, agricultural water demand could rise by as much as 60.6% even under a low-carbon scenario, while agricultural droughts run 35% longer if emissions are not cut, and restrictions grow ever more frequent. The reality is already visible: the Pyrénées-Orientales have been in drought for close to four years, with restrictions eased at the end of May 2025, reimposed within days as the June 2025 heatwave set in, and then extended by the prefect until at least 31 December as aquifer levels kept falling; the Canal du Midi closed to navigation two months early in autumn 2025, with reservoirs at 10% of capacity in November.12323334

6. Collapse cascades: when systems interact

The trap of feedback loops

Understanding collapse risk means grasping the feedback loops between systems. These are not linear collapses, but cascades:

1. Prolonged drought + heatwave → drying of soils → stressed vegetation.
2. Stressed vegetation → less evapotranspiration → fewer local clouds → even more heat (a positive feedback).
3. Dried-out soils → less infiltration → aquifers that fail to recharge the following winter → an even more stressful start to the next season.
4. Weakened forests → a windfall for wood-boring bark beetles → dead trees → ideal fuel for fires → additional CO₂ emissions → faster warming.

This is exactly what France has observed since 2018: a self-feeding cascade.352722

The memory effect of repeated droughts

Studies of European beech show that the trees which lost crown vitality in the 2018 drought had already been growing slowly for decades, and that growth recovered within two years only for those that lost less crown and only where the following years turned wet; trees with severe crown loss had still not recovered two years on. The repetition of 2003 and 2018–2019 in less than two decades has created a structurally weakened forest population, many of whose individuals enter 2026 with already degraded carbon and water reserves. For crops, INRAE confirms the same mechanism: poor-harvest years (2003, 2007, 2016, 2024 for wheat) result from a succession of extreme events, not necessarily a single one. The simultaneity of heatwave, drought, disease, and depleted soil becomes the trigger.362611

7. Tipping scenarios: a plausible chronology

Within 1 to 3 years (2026–2029)

Immediate (summer 2026). The warning signals are already red. If the current drought persists to September 2026 without significant rain, the 2026 maize harvests, which began in correct conditions but face temperatures above 40 °C during critical phases, could record decreases similar to or greater than 2003 (−30%); the slower-responding aquifers would keep falling into the autumn, to levels below those of 2025; forests already weakened by earlier droughts risk accelerated dieback; and fire risk is high, with a third of départements already on alert at the end of June.92238

Short term (2027–2029). Even with a rainy winter, climate trajectories point to a growing probability of extreme combinations. The 2026 INRAE study defines the “worst harvest” years for wheat as those falling 10% below the yield trend, and identifies 2003, 2007, 2016, and 2024 on that basis. Under a high-emissions scenario it projects that simultaneous heatwaves and droughts could become three times more frequent in northern France and six times more frequent in the south.11

The 2030–2040 window: the real systemic tipping threshold

This is the most critical decade according to converging projections. On the TRACC trajectory, metropolitan France reaches +2 °C nationally around 2030, when global warming reaches +1.5 °C, and +2.7 °C around 2050. It is in that first window that several major transitions become possible at once.1

• Forests. Beech decline already observed in central Europe and in north-eastern France extends to the beech-dominated zones of the southern half of the country. North-eastern spruce, on which bark beetles already kill some 2.4 million m³ a year, keeps losing ground. Oak dieback, so far localised and without epidemic dynamics, was noted more frequently in 2023 and 2024, with pedunculate oak worst affected. No source publishes a viability threshold for any of these, and the warming level at which each gives way is a reading across the evidence, not a published finding.
• Agriculture. Non-irrigated maize in the South-West could become economically unviable in some years; stone-fruit orchards face phenological disruption and late-frost risk; and the vine’s margin narrows too, with flowering and ripening arriving earlier and leaving it more exposed to spring frost, while warming is expected to reduce grape quality and production overall and to make regions such as Bordeaux less suited to viticulture.
• Water. A third of the population of southern Europe risks water shortage from +2 °C of global warming, which the same trajectory places around 2050, one decade beyond this window.

26252229303940414243

End of century (+4 °C in France): the irreversible

By 2100, under the TRACC scenario, projections converge toward irreversible transformations: a significant increase in tropical nights (nights that do not fall below 20 °C) across almost all of metropolitan France, with only mountainous areas, Brittany and the Channel coast remaining virtually unaffected; 40 to 50 such nights per year in the northern half and more than a hundred in the most exposed areas, notably around the Mediterranean; −19% summer precipitation on the national average, far worse in the South; summer soil moisture matching today’s extreme-drought conditions as the norm; southern France climatically similar to today’s Andalusia, the North to today’s southern France. Under such conditions, irrigated agriculture would become impossible in areas productive today, and current broadleaf forests would recompose toward more xerophilous formations (garrigue, maquis) or forests stripped of key species such as beech.21

8. What “collapse” means in concrete terms

Durations and practical thresholds

How long a drought or heatwave must last to cause vegetation collapse has no single answer, because it depends on the initial state (a soil already in deficit is far more vulnerable), the combination of factors (temperature, rainfall, aquifer state, and prior stress), and the ecosystem concerned. The ranges below are working rules of thumb for stress-testing a site. They are not thresholds established by any single study, and no source reviewed here reports them in this form; treat them as orders of magnitude to be checked against the system in front of you.

• Non-irrigated grasslands can brown and die after as little as 3 to 6 weeks of drought plus summer heat. They may regenerate in autumn, but the soil can stay degraded.
• Annual crops (maize, sunflower) during critical phases: above 35 °C for 5 to 10 consecutive days during pollination can cut yields by 30 to 60%; a 4-to-8-week drought during grain-filling can compromise the whole harvest.
• Forests die more slowly (over years), but the point of no return exists: a severe drought of 2 to 4 consecutive growing seasons (SPEI below −1.5) can drive irreversible dieback.

The interactive tool below places a chosen duration against these ranges. For forests the two-to-four-season band is indicative and not a measured threshold: the Swiss beech monitoring behind it followed trees for three seasons after 2018 and found dieback still partly reversing in the third, which its authors judged too short a window to fix where recovery stops. The underlying literature on drought damage is cited alongside: SPEI-based damage thresholds for grassland and cropland15, the 2003 heatwave outcomes for European crops28, drought and heatwave effects on Western Mediterranean vegetation 21, and beech decline after 2018 in Switzerland25 and north-eastern France 26.

Cascading versus immediate collapse

It helps to distinguish two types. Acute collapse (a record heatwave like 2003 or 2026) causes immediate, massive losses (scorched harvests, human mortality, fires); it is visible and painful, but systems can regenerate if it is not repeated. Chronic collapse is the repetition of extreme events at a frequency too high to allow recovery between episodes: the process French forests have been living since the 2018 drought, and which the IGN measures across its two inventory windows, mortality rising from 7.4 Mm³ per year over 2005–2013 to 16.7 Mm³ per year over 2015–2023, and sequestration falling from 63 to 39 MtCO₂ per year over the same windows. It is not spectacular; it is a silent slide toward a degraded new norm.22

Irreversible tipping is the point where ecological recovery is outpaced. Model-data studies of drought-induced tipping points show that several consecutive dry years can drain soil moisture and tree carbon reserves far enough for productivity to collapse instead of recovering. In a Californian mixed conifer forest, that followed four years at 45% below normal precipitation, and adding 157 mm, some 15% of the annual average, in the driest year was enough to steer the stand back toward recovery. No equivalent threshold has been published for French lowland forests, but the Mediterranean rim and the Midi are where the same conditions are approached first.1826

9. Socio-economic impacts: the cost of inaction

Economic projections poorly integrate cascading collapses, but they still convey the stakes. Without action, climate losses in France could reach 5.82% of GDP in 2100 (versus 1.81% by limiting warming to +2 °C); by 2050, annual GDP losses could reach up to 30 billion euros even on a low-carbon pathway; and as far back as 2019 the Fédération française de l’assurance put the annual cost of agricultural drought at up to 700 to 900 million euros, varying from year to year and rising.1244

In 2026, the May and June heatwaves struck a winter wheat crop rated 80% good to very good as of 18 May, with a documented risk of sharply reduced pollen viability above roughly 31 °C in the days before flowering, and the summer maize crops are entering their critical phases in the worst conditions since 2003. Modelling by ADEME economists with the ThreeME model, published by INSEE in Economie et Statistique, puts the cost of climate inaction for France at nearly seven percentage points of annual GDP by 2100 relative to an orderly transition that limits warming to 1.5 °C, with the trajectories diverging from about 2030 and the damage building steadily from then on.204537

10. Conclusion: toward collective clear-sightedness

France is not facing a hypothetical future risk. It is living, in real time, the early stages of a gradual ecological and agricultural collapse. The data are convergent and unequivocal: forest mortality has doubled in ten years and French forests now absorb 39 MtCO₂ per year, 38% less than the 63 MtCO₂ per year of the preceding decade; each drought leaves ecosystems weaker than before; the current trajectory (France at +4 °C in 2100) will bring conditions structurally incompatible with current agriculture in the South and a profound recomposition of forests; the 2030–2040 decade, in which France passes +2 °C nationally, is the window this article judges most likely to see several systems give way at once; and the summer of 2026 is a concrete alarm: the hottest day ever recorded, on deficit soils, after the hottest spring since 1930.242221653

Collapse thresholds are not precise dates but corridors of increasing probability. What the science indicates clearly is that French natural and agricultural systems are today crossing thresholds of resistance that, cumulatively, could set off irreversible transitions well before 2050. Adaptation is no longer a preventive option: it is an immediate necessity, one that must accept that part of the damage is already irreversible and prepare for a fundamentally different future in terms of biodiversity, food production, and water management.

References

• 1.Consequences for France // Climate change observations (SDES) // Average warming of 2 °C in 2030, 2.7 °C in 2050 and 4 °C in 2100 for metropolitan France.
• 2.What will France's climate be like at +4 °C? (Le Monde) // Météo-France forecasts for temperature and precipitation.
• 3.France records its hottest day with relief on the horizon (Le Monde) // A day–night average of 30 °C set a new heat record on 24 June 2026.
• 4.Europe heatwave breaks records as UN agencies ramp up health warnings (UN News) // A record-breaking heatwave grips large parts of Europe (WMO).
• 5.France : après un printemps des records, l'été sera probablement encore chaud // National anomaly of +1.7 °C vs the 1991–2020 normal; hottest spring since 1930.
• 6.Pourquoi l'épisode de chaleur de fin mai est-il inédit par sa précocité (Météo-France) // A durable, intense heat episode between 21 and 30 May 2026.
• 7.Western Europe has hottest June on record (WMO) // France recorded its hottest day on record on 24 June, reaching 43.8 °C at Pulluau. WMO news, 9 July 2026.
• 8.Nappes d'eau souterraine au 15 juin 2026 (BRGM) // Aquifer drawdown accelerates, with 86% of levels declining.
• 9.Groundwater levels on 1 June 2026 (BRGM) // Water levels dropped in 77% of aquifers after the April–May rainfall deficit.
• 10.Climate 2050: what can we expect? (CircularPlace) // Regional climate analogues for France by 2050.
• 11.Wheat production: anticipating climate risks in France up to 2100 (INRAE) // Rare extreme events become common in future projections for French wheat.
• 12.France // G20 Climate Risk Atlas (CMCC, 19 October 2021) // Country factsheet with 2050 projections for heatwaves and water demand, and GDP losses for 2050 and 2100.
• 13.France prepares for another 'very intense and widespread' heat event (Le Monde) // Southern half of France hit first; temperatures possibly over 40 °C.
• 14.Global variations in critical drought thresholds that impact vegetation (PDF, MPG)
• 15.A multi-metric assessment of drought vulnerability across vegetation types (PubMed) // High-resolution remote sensing of drought impact across vegetation types.
• 16.Flash droughts exacerbate global vegetation loss and delay recovery (Nature Communications) // NDVI loss during flash droughts (9.0%) is ~1.5× higher than conventional droughts.
• 17.Future Drought Will Lead to a Decrease in Vegetation Resilience in China (Earth's Future) // GPP resilience time of 1–9 months.
• 18.Forest productivity recovery or collapse? Drought-induced tipping points (Global Change Biology) // Model–data integration on drought-induced forest tipping points.
• 19.The prevalent life cycle of agricultural flash droughts (npj Climate and Atmospheric Science) // Characteristics and life cycle of agricultural flash droughts globally.
• 20.France Heat Wave Threatens Wheat Crop (Lukas Graf, LinkedIn) // Heat wave peaking over France and neighbouring Europe.
• 21.Droughts and heatwaves in the Western Mediterranean: impact on vegetation and wildfires using the coupled WRF-ORCHIDEE regional model (RegIPSL) // Climate Dynamics // Combined drought and heatwave raise burned area 2.1× and fire radiative power 2.9×; drought alone reduces LAI by 10% on average and up to 23% locally.
• 22.Résultats 2025 de l'Inventaire forestier national (IGN) // Mortality up 125% in ten years; dead trees now 5% of standing volume.
• 23.Forêts françaises : la mortalité des arbres s'envole, le puits de carbone ralentit (We Demain) // Drought, heat, disease and pests degrade French forest health.
• 24.Forêt : une mortalité en hausse à cause du changement climatique (Vie-publique) // Tree mortality has doubled in ten years: from 7.4 to higher annual volumes.
• 25.European beech dieback after premature leaf senescence during the 2018 drought (PMC) // Widespread premature leaf senescence during the 2018 hot drought in northern Switzerland.
• 26.Impacts of recurrent extreme drought on beech, 2013–2022, NE France (EGU24)
• 27.Beech decline literature (ProQuest) // Scholarly record on recurrent drought and beech.
• 28.European heatwave impact on the agricultural sector (IPCC AR4 WGII Chapter 5, Box 5.1, PDF) // Winter wheat near maturity suffered less; summer crops far more.
• 29.Agriculture française en 2050 : marché ou résilience territoriale ? (Banque des Territoires, SCET) // 16 of 24 crops strongly vulnerable; 42% of farmland threatened.
• 30.Analysis of climate change impacts on EU agriculture by 2050 (JRC PESETA IV, PDF) // Near-collapse of European maize production projected around 2050.
• 31.Accelerating the climate transition with a low-carbon, resilient and fair food system // Executive summary (Haut Conseil pour le Climat, January 2024, PDF) // Major crop losses and reduced livestock feed resources, with further yield losses for summer crops at around +2 °C by 2050.
• 32.Heatwave brings more water restrictions across France (Connexion) // Areas already limiting water use.
• 33.Canal du Midi closed to boats until March due to low water levels (Connexion) // Extended drought in the south of France.
• 34.Water restrictions extended in Pyrénées-Orientales as drought continues (Connexion) // Groundwater levels continue to fall.
• 35.Impacts on and damage to European forests from the 2018–2022 heat and drought events (NHESS, Copernicus) // Highly significant crown thinning in France, about 19 million m³ of spruce lost to bark beetles in its two worst-hit regions, and a record burnt area in 2022.
• 36.Slower growth before 2018 predisposed European beech to crown dieback (Science of the Total Environment)
• 37.Late-May Heat Wave May Stress France Wheat (DTN/Progressive Farmer) // Extreme late-spring heat in Europe is becoming more frequent.
• 38.France on red alert for forest fires amid extreme heat (FRANCE 24) // Record-breaking heatwave with a third of départements on fire alert.
• 39.Dépérissement des forêts : 3 essences menacées (ONF) // State of forest dieback in France.
• 40.Santé des forêts (HAL, PDF) // Localised but increasingly frequent dieback noted in 2023–2024.
• 41.Rapport du Giec sur le climat : et la France dans tout ça ? (L'info durable) // Agricultural and viticultural losses, deadly heatwaves, flooding.
• 42.Giec : les pronostics alarmants sur le réchauffement climatique en France (Elle) // IPCC forecasts for France.
• 43.France: Country Water Resources (Climate Scorecard) // Water restrictions are now a regular tool for managing shortages.
• 44.France's path towards sustainable farming eroded by increased droughts (Euractiv) // Intense drought has already eliminated fallow fields.
• 45.Macroeconomic Impact of Climate Damage in France (Jacquetin and Callonnec, ADEME, in Economie et Statistique 543, INSEE, PDF) // ADEME's ThreeME model puts climate inaction at close to 7 points of annual GDP by 2100, against an orderly transition that limits warming to 1.5 °C.

Sources: TRACC/SDES 2024, Météo-France, BRGM (June 2026 bulletins), IGN National Forest Inventory 2025, INRAE 2026, JRC PESETA IV, G20 Climate Risk Atlas France, Le Monde, WMO. Translated from the original French research notes; figures preserved as published.